Nobody has ever looked at whether a KMT2A fusion is at PML nuclear bodies - a counted zero, with the one biochemical hint pointing yes
also called the KMT2A fusion PML-body question · ND10 and the fusion
Cell nuclei contain small distinct compartments called PML bodies, and for one other leukaemia fusion protein their behaviour is the whole story of how the disease is treated. For this fusion, no one has ever put the two in the same microscope field - and the one biochemical experiment that exists says the fusion and PML do come out of a cell together.
The same thing, for a scientist
**The literature is empty and the emptiness is counted.** As of 2026-09-24, `"PML nuclear bodies"[tiab] AND MLL[tiab]` returns 0, the same against KMT2A returns 0, `"PML nuclear body"[tiab] AND MLL[tiab]` returns 1 (an autophagy review, off-topic on title check) and `ND10[tiab] AND MLL[tiab]` returns 0, against operand controls of 644 for the bare phrase and 6,329 for bare MLL [E1]. **And there is a hint, in a supplementary table, pointing the other way from the absence.** PML is recovered against **all seven** MLL-fusion baits in the only AP-MS survey of these proteins, including MLL-AF4 - and it is the *only* PML-body component in the 960-protein network, with SP100, RNF4 and DAXX all absent [E2]. So the biochemistry says the fusion and PML co-purify; the cell biology has never been done. **Why this is a priority rather than a curiosity.** This graph already holds `localization-pml-nuclear-bodies-pmlrara`, because for PML::RARA the subnuclear address *is* the disease mechanism and the microspeckled immunofluorescence pattern was diagnostic for a decade before anyone understood it. Arsenic trioxide works by degrading PML and reorganising those bodies. A two-colour immunofluorescence of a KMT2A fusion against PML in a t(4;11) line is an afternoon's work in any lab that has the antibodies, and the result is currently unpredictable in either direction. **Absence class: `never-measured`.** Not `measured-not-reported` - there is no imaging deposit to reanalyse, because no one has taken the images.
The queries, with their translations, so the absence can be audited
| query | records | querytranslation |
|---|---|---|
"PML nuclear bodies"[tiab] | 644 | "PML nuclear bodies"[Title/Abstract] |
MLL[tiab] | 6,329 | "MLL"[Title/Abstract] |
ATRX[tiab] | 2,103 | "ATRX"[Title/Abstract] |
"PML nuclear bodies"[tiab] AND MLL[tiab] | 0 | both terms, tagged |
"PML nuclear bodies"[tiab] AND KMT2A[tiab] | 0 | both terms, tagged |
"PML nuclear body"[tiab] AND MLL[tiab] | 1 | off-topic on title check: The role of autophagy in targeted therapy for acute myeloid leukemia |
ND10[tiab] AND MLL[tiab] | 0 | both terms, tagged |
Operand controls are large on both sides, so this is not the two-large-sets-with-an-empty-intersection failure this project has documented. Every query was run one tag per term, never as a parenthesised (A OR B)[tiab] group, and every querytranslation was read back.
A trap found while running this, worth recording. hDaxx[tiab] AND leukemia[tiab] returns 19 records, which looks like a DAXX-leukaemia literature. All nineteen were enumerated and title-checked and not one is about a leukaemia. Every hit is herpesvirus or cytomegalovirus ND10 biology, and the word leukemia is present because PML's full name is promyelocytic leukemia protein. In any query about this compartment, leukemia is a contaminated token.
What could be measured, in order of cost
- Two-colour immunofluorescence, KMT2A N-terminal antibody or a tagged fusion against PML, in SEM or RS4;11, with a PML::RARA-positive NB4 control for the microspeckled pattern and an untransformed lymphoblastoid line for normal bodies. Hours. Answers the question as posed.
- Count and size the bodies under menin inhibition and under fusion degradation.
gene-pmlalready asks this. It converts a static localisation into a perturbation experiment at no extra reagent cost. - Test whether the AP-MS PML signal is SUMO-dependent by repeating the pulldown with a SUMOylation-deficient bait or a PML SIM mutant. This is the control that separates a real contact from a SUMO-abundance artefact, and
gene-pmlnames it. - Arsenic trioxide on a KMT2A-rearranged line. ATO is approved, cheap and dissolves PML bodies. Whatever it does to a t(4;11) blast is currently unknown and unpredictable, and the experiment is a plate reader away.
None of these needs a new reagent, an animal or a patient sample.
What this entry deliberately does not claim
It does not claim the fusion is at PML bodies. One co-purification row in a supplementary table, from a protein famous for appearing in affinity purifications it has no business in, is not a localisation. The claim is precisely: the question has never been asked, the one relevant biochemical observation is positive, and the experiment is cheap.
It also does not claim that DAXX is at PML bodies in a leukaemic blast. DAXX's PML-body residence is established in other cell types and is a cell-cycle-dependent equilibrium with heterochromatin rather than a fixed address; nothing here is a measurement in this disease.
| compartment | PML nuclear bodies (ND10, PML oncogenic domains) |
|---|---|
| subject | KMT2A fusion proteins |
| localization shown for fusion | NO - nobody has looked in any fusion. Zero records in titles and abstracts for the compartment against MLL or KMT2A; the only relevant observation is a co-purification of PML with all seven MLL-fusion baits in one supplementary table. |
| absence class | never-measured |
| what closes it | Two-colour immunofluorescence of a KMT2A fusion against PML in SEM or RS4;11, with NB4 (PML::RARA) as the microspeckled-pattern positive control. Hours of work, no new reagents. |
| relationship to existing entries | `localization-pml-nuclear-bodies-pmlrara` covers the compartment for the OTHER leukaemia fusion, where it is the disease mechanism, and asks whether a subnuclear pattern would have diagnostic value for KMT2A fusions. `gene-pml` asks whether the fusion localises there. Neither bounds the absence, gives it an absence_class or records the queries; this entry does those three things and adds the co-purification hint and the token collision. |
queries run — the numbers behind the chart
[
{
"query": "\"PML nuclear bodies\"[tiab]",
"count": 644,
"role": "operand control"
},
{
"query": "MLL[tiab]",
"count": 6329,
"role": "operand control"
},
{
"query": "\"PML nuclear bodies\"[tiab] AND MLL[tiab]",
"count": 0,
"role": "primary"
},
{
"query": "\"PML nuclear bodies\"[tiab] AND KMT2A[tiab]",
"count": 0,
"role": "primary"
},
{
"query": "\"PML nuclear body\"[tiab] AND MLL[tiab]",
"count": 1,
"role": "primary, off-topic on title check"
},
{
"query": "ND10[tiab] AND MLL[tiab]",
"count": 0,
"role": "primary"
},
{
"query": "hDaxx[tiab] AND leukemia[tiab]",
"count": 19,
"role": "alias probe - ALL 19 off-topic, the token 'leukemia' coming from 'promyelocytic leukemia protein'"
}
]name list searched
[ "PML nuclear bodies", "PML nuclear body", "ND10" ]
name list not searched
[ "PML oncogenic domains", "POD", "nuclear domain 10", "PML-NB", "the fusion partner symbols in place of MLL/KMT2A" ]
token collision found
{
"token": "leukemia",
"collides_with": "the word 'leukemia' inside 'promyelocytic leukemia protein' (PML)",
"measured": "hDaxx[tiab] AND leukemia[tiab] returns 19 records, 19 of 19 off-topic on title check, all herpesvirus/CMV/adenovirus/BK-virus ND10 biology",
"affects": "any query about the PML compartment scoped by the word leukemia, including PML, SP100 and ND10 queries",
"registry_status": "NOT in data/symbol-collisions.json as of 2026-09-24; the existing PML row covers the gene/disease/neurological-disease ambiguity, not this one. Reported to the coordinator for addition; not edited by this beat."
}measurements
[
{
"quantity": "pubmed_record_count",
"subject": "\"PML nuclear bodies\"[tiab] AND MLL[tiab]",
"value": 0,
"units": "records",
"n": null,
"universe": "PubMed titles and abstracts, 2026-09-24",
"null": 644,
"note": "null column is the bare-phrase operand control; bare MLL is 6,329, so this is not an empty intersection of two small sets"
},
{
"quantity": "off_topic_fraction",
"subject": "hDaxx[tiab] AND leukemia[tiab], all hits title-checked",
"value": 1,
"units": "fraction",
"n": 19,
"universe": "PubMed titles and abstracts, 2026-09-24",
"null": null,
"note": "19 of 19 off-topic; the apparent DAXX-leukaemia literature is PML's full name"
}
]- If the fusion does sit at or near PML bodies, arsenic trioxide becomes an obvious thing to try - and the direction of its effect is not predictable. ATO degrades PML; if the bodies are restraining the fusion rather than hosting it, dissolving them could make the disease worse, and there is no data either way.
- A negative immunofluorescence would not close the question. PML bodies are dynamic and DAXX's own residence there is a cell-cycle-dependent equilibrium with heterochromatin, so a single fixed-cell image at an unsynchronised time point can miss a real association.
- Do the immunofluorescence. A KMT2A fusion against PML in SEM or RS4;11 with an NB4 control is hours of work and the question has stood open since the compartment was described.
- Is the AP-MS PML signal SUMO-dependent? A SUMOylation-deficient bait or a PML SIM mutant separates a real contact from a SUMO-abundance artefact, and the single weak SUMO2 row in the same table is the only evidence either way.
- What does arsenic trioxide do to a KMT2A-rearranged blast? ATO is approved and dissolves PML bodies; the experiment is a plate reader away and the answer is unpredictable from anything in this graph.
- Search the full text - not the abstracts - of the papers that have imaged a tagged KMT2A fusion. A co-localisation panel in a supplement would refute this absence, and abstract indexing cannot see one.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
As of 2026-09-24, PubMed returns 0 records for '"PML nuclear bodies"[tiab] AND MLL[tiab]', 0 for the same against KMT2A[tiab], 0 for 'ND10[tiab] AND MLL[tiab]' and 1 for '"PML nuclear body"[tiab] AND MLL[tiab]' which is off-topic on title check, against operand controls of 644 records for the bare phrase and 6,329 for bare MLL[tiab]; all querytranslations were non-empty and correctly tagged.
Bounded PubMed absence for PML nuclear bodies against MLL and KMT2A, run through scripts/pubmed_safe.py with operand controls and title checking of every hit · PubMed (NCBI E-utilities), via this project's pubmed_safe wrapper 2026What this evidence cannot show
Titles and abstracts only. Phrases tried: 'PML nuclear bodies', 'PML nuclear body', 'ND10'. NOT tried: 'PML oncogenic domains', 'POD', 'nuclear domain 10' spelled out, 'PML-NB', or the fusion-partner symbols in place of MLL/KMT2A. Per rule 70 a compartment absence is bounded by its name list, and this list is four strings. A colocalisation shown only in a supplementary immunofluorescence panel is invisible to an abstract index - which is exactly the failure mode this project documented for an H3K79me2 ChIP in Figure 6A - so the honest bound is 'zero in titles and abstracts; supporting figures not searched'.
Evidence that argues against this (1)
- A co-localisation is precisely the kind of result that appears as one supplementary panel and never reaches an abstract, so an abstract-bounded zero is weak for this class of claim specifically. The strongest available refutation would be a full-text search of the dozen or so papers that have imaged a tagged KMT2A fusion at all; that was not run, and until it is, this absence bounds the indexed literature rather than the field. PubMed (NCBI E-utilities) 2026
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
PML is recovered against all seven MLL-fusion baits in the only AP-MS survey of MLL fusion proteins - MLL-AF4 at p 0.0282 with 4 and 4 spectral counts, MLL-AF9 at p 6.61e-06 with 22 and 22 - and it is the only PML-nuclear-body component in the 960-protein network, with DAXX, SP100 and RNF4 all absent.
MLL-fusion-driven leukemia requires SETD2 to safeguard genomic integrity. · Skucha A, Ebner J, Schmöllerl J, et al. · Nature Communications 2018What this evidence cannot show
Read directly from the cached Supplementary Table 5 on 2026-09-24; the paper's text does not discuss PML and the phrase 'PML nuclear bodies' does not appear in it, which is why this observation is filed as a bounded-absence entry rather than as a positive localisation. A co-purification in a lysate is not a subnuclear address. PML is abundant, SUMO-rich and a frequent passenger in affinity purifications.
Evidence that argues against this (1)
- The absence of every other body component is as consistent with 'PML is sticky in AP-MS' as with 'the fusion contacts PML directly'. Recovering a scaffold without its clients is also what a compartment dissolved by the lysis buffer looks like. This item is therefore a reason to do the immunofluorescence, not a substitute for it, and an entry that treats it as evidence of localisation has overread a single supplementary row. Ishov AM et al. 1999
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
A naive alias search suggesting a DAXX-leukaemia literature is an artefact of PML's full name: 'hDaxx[tiab] AND leukemia[tiab]' returns 19 records, all 19 were enumerated and title-checked on 2026-09-24, and every one is herpesvirus, cytomegalovirus, adenovirus or BK-virus ND10 biology in which the word leukemia appears only as part of 'promyelocytic leukemia protein'.
Enumeration and title check of all 19 records returned by hDaxx[tiab] AND leukemia[tiab] · PubMed (NCBI E-utilities), titles fetched by efetch and read individually 2026What this evidence cannot show
All 19 titles were read; none of the 19 abstracts was read, so a paper whose title is about herpesvirus and whose content includes a leukaemia experiment would be missed. The collision is with the token 'leukemia' inside 'promyelocytic leukemia protein', not with hDaxx itself, so it affects any query about this compartment scoped by the word leukemia - including queries about PML, SP100 and ND10.
Evidence that argues against this (1)
- The collision cuts the other way too, and this entry should not be read as saying DAXX has no leukaemia biology. The abstract of the Daxx haematopoietic knockout paper describes Daxx as 'a retrotransposable element repressor INACTIVATED IN MYELOID LEUKAEMIA and other neoplasms' - so there is a DAXX-leukaemia literature, it is about LOSS of DAXX, and it points in the opposite direction from the dependency this beat's DepMap analysis found. Gerber JP et al. 2021
the same compartment, where for PML::RARA the subnuclear address IS the disease mechanism and here it has never been looked at
bounds the absence gene-pml's open question names, and adds the 7-of-7 co-purification and the 'leukemia' token collision
- related to →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
the subnuclear address of the index fusion has never been measured against this compartment
- analogous to →PML::RARA, t(15;17)(q24;q21)
the comparator fusion for which this compartment is diagnostic, mechanistic and druggable with arsenic
records the hDaxx/leukemia token collision - 19 of 19 hits off-topic - and the opposite-direction DAXX-leukaemia literature the beat's framing missed
the 7-of-7 PML co-purification is the one biochemical hint behind this otherwise empty question